McGinleyDynamic — McGinley Dynamic

An average that adjusts its own speed to the gap between itself and price: the further price runs, the harder the line chases. Designed to avoid the whipsaw of a fixed-period average when the market changes pace.

Signature

McGinleyDynamic(period, on=<source>)

Argument	Type	Required	Notes
period	int Param or literal	Yes	Base speed. Default 14, bounds 1–400.
on=	symbol, OHLCV field, or indicator	No	Source. Defaults to the trade symbol’s close.

Formula

$$ MD_t = MD_{t-1} + \frac{C_t - MD_{t-1}}{N,(C_t / MD_{t-1})^4} $$

The fourth power is what makes the response asymmetric: it accelerates hard when price runs away from the line.

Warmup

Starts at the first close; no window to fill.

Reading the value

md.value
self.trade.close > md.value

Example

class McGinleyTrail(Strategy):
    """Follow the dynamic line; it tightens automatically when price accelerates."""
    period = Param(14, min=1, max=400)
    trade = Symbol()
    sizing = FixedCapital()

    def execute(self):
        md = McGinleyDynamic(self.period, on=self.trade)
        if self.trade.close > md.value and self.trade.is_flat:
            self.buy(self.trade, reason="Above McGinley")
        if self.trade.close < md.value and self.trade.is_long:
            self.sell(self.trade, self.trade.position, reason="Below McGinley")

Notes

• John R. McGinley, Journal of Technical Analysis.
• Proven by an assert-via-trade run on SPY daily (matched an independent strategy-side recursion within 0.05; zero alarm trades).
• Guarded against a non-positive line value (it would make the ratio meaningless); on that bar the line resets to price.
• A different recipe from pandas_ta’s mcgd(10), not a different scale: measured on a 300-bar tape ours stays about 0.7 % away and never converges to it, because the construction (window offset, seed, weight normalisation) is its own.